Perplexity Research

---
name: perplexity-research
version: 0.1.0
description: Brain-augmented Perplexity research. Pulls brain context about a topic, then drives Perplexity (the agent's Perplexity tooling or API; any web-search tool as fallback) with that context so the web pass surfaces what is NEW vs what the brain already knows, with citations. Use for entity enrichment, current-state checks, deal monitoring, and freshness deltas. NOT for simple URL fetches (plain fetch) or brain-only questions (use the search tool).
triggers:
  - "Perplexity research"
  - "perplexity-research"
  - "perplexity"
  - "what's new about"
  - "current state of"
  - "what changed about"
  - "surface new developments"
mutating: true
writes_pages: true
writes_to:
  - research/
---

# perplexity-research — Brain-Augmented Perplexity Research

> **Convention:** see `conventions/quality.md` (via `get_skill conventions/quality`)
> for citation rules; every claim from Perplexity research lands with a verifiable
> citation, not a paraphrase.
>
> **Convention:** see `conventions/brain-first.md` (via `get_skill conventions/brain-first`)
> for the lookup chain. This skill ENFORCES brain-first by loading brain
> context BEFORE any web call — the web search focuses on the delta
> between brain knowledge and current web state.

## What this does

Combines existing brain knowledge with the agent's web-search tooling.
The agent first pulls what the brain already holds on a topic, then runs
web searches and page reads framed by that context, synthesizing multiple
sources with citations, focused on what's NEW relative to the brain.

**The key insight:** a research pass is only useful if it knows what you
already know. By loading brain context first and searching against it,
the output is the delta — not a re-narration of settled fact.

## When to use this vs other tools

| Need | Use |
|------|-----|
| Deep research with citations | **This skill** — brain context + web search + synthesis |
| Quick URL content | the agent's plain page-fetch tool |
| Brain-only lookup | `search` (MCP) / `memrain search` |
| Real-time social monitoring | external social-media collectors |
| Structured data lookup against a tracker | `skills/data-research/SKILL.md` |

## Output structure

The research output lands as a brain page under `research/<slug>` with
this structure:

```markdown
---
title: "[Topic] — Research [YYYY-MM-DD]"
type: research
date: YYYY-MM-DD
brain_context_slugs: ["pages whose context framed the web pass"]
recency_filter: "[hour|day|week|month|none]"
---

# [Topic] — Research [YYYY-MM-DD]

> Executive summary: 2-3 sentences on the delta between brain knowledge
> and current web state.

## Key New Developments
What's changed since the brain was last updated on this topic.

## Confirming Signals
Web evidence validating existing brain knowledge.

## Contradictions or Updates
Things that conflict with the brain — these need a closer look.

## Recommended Brain Updates
Specific page updates the user might want to make based on this research.
Each item: which page, what to add or change, source URL.

## Citations
- [Source title](URL) — accessed YYYY-MM-DD
- [Source title](URL) — accessed YYYY-MM-DD
- ...
```

## Invocation

The skill is markdown agent instructions; the agent uses its own
web-search and page-fetch tooling for the web leg:

```
# 1. Pull brain context
page_get <slug>                      # or
search "<topic keywords>"            # plus entity_facts / entity_timeline for entities

# 2. Frame the web pass with brain context inline:
#    """
#    Topic: <topic>
#    What we already know: <embedded brain content>
#    Find: what's NEW since 2026-MM-DD that the brain doesn't reflect.
#    Cite every claim.
#    """
#    Run the agent's web-search tool with queries derived from this frame;
#    fetch and read the top sources, not just their snippets.

# 3. Write the structured research page:
page_put research/<slug>

# 4. Cross-link entities mentioned (people, companies) per Iron Law:
link <entity-slug> <research-slug>
```

## Depth routing

| Depth | Cost profile | Use when |
|-------|-------------|----------|
| Deep pass (many searches + full page reads + synthesis) | slower, thorough | Entity enrichment, deal research, contradiction hunting |
| Quick pass (1-2 searches, snippet-level) | fast, cheap | Quick lookups, bulk monitoring, briefing pipelines |

Default to the deep pass. Drop to the quick pass for bulk / scheduled
contexts where turnaround matters more than depth. (Server-side model
routing is not involved here — the web leg and synthesis run in the
agent; the brain's Bedrock tiers only serve retrieval-side utilities.)

## Integration patterns

### Entity enrichment

Called by `skills/enrich/SKILL.md` when an entity page (person, company)
needs current web context:

```
page_get people/<slug>
# Use the page content as the "what we already know" frame, run the web
# pass for current news / role / context, then update the brain page
# with what's new (page_append or page_put).
```

### Deal / company monitoring (scheduled)

For each active item under `deals/` or `companies/`:

```
# Weekly (systemd timer or the agent harness's scheduler): pull recent
# news per company; flag changes for review as research/ pages.
```

### Morning briefing

Replace raw page-fetch calls in briefing pipelines with this skill so
the agent doesn't re-narrate already-known facts.

## Recency filter

Constrain the web pass by recency when the tooling supports it:
`hour | day | week | month`. Useful for news-cycle topics; omit for
evergreen research. Record the filter used in the page frontmatter.

## Anti-Patterns

- ❌ Sending NO brain context. Then it's just a search — use a plain
  fetch instead.
- ❌ Truncating the brain context. The whole point is "knows what you
  know." Load dense context.
- ❌ Discarding citations. Every claim in the output must have a URL.
- ❌ Skipping the cross-link step when entities are mentioned. Iron Law.

## Environment

- Requires the agent harness's web-search / page-fetch tools to be
  available. No server-side API key: the brain never calls the web.

## Related skills

- `skills/academic-verify/SKILL.md` — wraps perplexity-research for
  citation-verified academic claim checking
- `skills/enrich/SKILL.md` — calls perplexity-research as part of the
  entity-enrichment loop
- `skills/data-research/SKILL.md` — structured-data trackers (different
  shape: parameterized YAML recipes, not free-form research)


## Contract

This skill guarantees:

- Routing matches the canonical triggers in the frontmatter.
- Output written under the directories listed in `writes_to:` (when applicable).
- Conventions referenced (`quality.md`, `brain-first.md`, `_brain-filing-rules.md`) are followed.
- Privacy contract preserved: no real names, no host-specific filesystem path literals, no external-project references.

The full behavior contract is documented in the body sections above; this section exists for the conformance test.

## Output Format

The skill's output shape is documented inline in the body sections above (see "Output", "Brain page format", or equivalent). The literal section header here exists for the conformance check.